Leave exacta betting at the racetrack, wrote Cliff Asness in his latest blog post.

The AQR founder argued that it is difficult enough to be right once, let alone twice, making exacta-style investing—gambling on two separate, but related outcomes, such as which horses will place first and second in a race—more risky than it is rewarding.

“This type of two-step bet is usually a bad idea,” Asness wrote. “At the very least, it is worse than the original, simpler, one-step idea.”

“It’s hard enough to be right. It’s much harder to be right multiple times in a row.” —Cliff Asness, AQR Asness defined the two-step bet as when investors forecast a specific event—inflation, for example—and then invest in companies based on which they believe will benefit or decline, should that forecast come true. Rather than buy or sell currency, these investors buy or sell stocks based on how companies might perform should a currency go up or down.

This means that in order to profit, two things need to occur: First, the forecasted event must come to pass. Second, stocks have to react the way the investor predicted.

By hinging a bet on two separate events, Asness argued, investors face a much lower probability of winning than if they had simply bet on the first event.

“It’s hard enough to be right,” Asness wrote. “It’s much harder to be right multiple times in a row.”

At least in horse racing, Asness added, gamblers are promised a hefty payout to compensate for the decreased probability of winning. In the financial markets, however, there’s rarely a bonus for taking on extra risk.

“If, somehow, you are really sure where the dollar is going, or where interest rates will be,” Asness concluded, “place your bet on the dollar or interest rates.”

Related: Asness: This Is Why Factor Investing Will Survive